Written by David Diez, Mine Çetinkaya-Rundel, and Christopher Barr, this open textbook offers a fresh take on the traditional introductory statistics curriculum. It is used in courses at institutions ranging from community colleges to the Ivy League, and it works well for self-study too.
The fourth edition brought a complete visual redesign and a reworked structure. Proportions are now used to introduce the core ideas of inference, so inference for categorical data comes before inference for numerical data, and two-sided hypothesis tests are used throughout. Many older datasets were replaced with newer ones, two multiple regression case studies were added to the main text, and exercises now appear at the end of each section, with review exercises closing each chapter.
The book covers:
- Collecting data through sampling and experiments
- Summarizing numerical and categorical data
- Probability, conditional probability, and random variables
- Normal, binomial, Poisson, and other common distributions
- Confidence intervals and hypothesis testing
- Inference for proportions and contingency tables
- Inference for means, paired data, and ANOVA
- Linear, multiple, and logistic regression
Beyond the text, OpenIntro provides videos, slides, and labs for each chapter, with labs available in R and several other statistical software packages. The PDF is free, low-cost paperbacks are available in black and white or color, and the book has been translated into Japanese and Uzbek.
Find out more about the book and the supplementary resources at openintro.org/book/os.